Keywords

1 Introduction

Sexual harassment is a universal phenomenon. According to the survey of the European Union Agency for Fundamental Rights, 45%–55% of women in the workforce in the European Union have experienced at least one form of sexual harassment (Latcheva 2017). Despite a high number of occurrences in the workplace, there is no consensus on a universally accepted definition of sexual harassment. MacKinnon (1979) underlined the absence of a “… generalized, shared, and social definition of [sexual harassment]” (p. 27). The absence of a universal definition poses a problem when targets of sexual harassment try to identify it and when strategies for preventing or dealing with sexual harassment are proposed (Uggen and Blackstone 2004). It is generally stated that an agreed upon more generalized and operational definition of sexual harassment would be helpful to develop mechanisms to prevent such behavior (Fitzgerald 1990). Barr (1993) argued that when individuals have different definitions and perceptions about sexual harassment, it becomes difficult to determine whether to file a complaint, which results in under-reporting of such incidents in the workplace. In the 1970s, sexual harassment received socio-legal recognition, which somewhat alleviated the problem from going unnoticed in North America (MacKinnon 1979).

In Europe, sexual harassment and its prevention attracted attention in the 1980s. However, sexual harassment incidences are generally under-reported. One reason for the under-reporting of sexual harassment may be the lack of a shared understanding of what constitutes sexual harassment.

It might be very challenging to establish a universal definition of sexual harassment given the cross-cultural differences in the behavioral standards of different countries (Barak 1997). Therefore, trying to establish a universal definition may not be feasible given the differences in cultural values related to behaviors constituting sexual harassment. Despite these difficulties, a widely used definition was offered by MacKinnon (1979) who states that sexual harassment is any “unwanted imposition of sexual requirements in the context of a relationship of unequal power” (MacKinnon 1979, p. 1). However, it should be stated that, from a practical perspective, this broad definition may be ineffective when trying to identify cases of sexual harassment. Recently, Latcheva (2017) indicated that sexual harassment is discrimination on the grounds of gender and argues that it violates equal treatment for both genders principle.

From a more practical standpoint, the Equality Act of 2010 defined sexual harassment as “unwanted conduct of a sexual nature which has the purpose or effect of violating someone’s dignity, or creating an intimidating, hostile, degrading, humiliating or offensive environment for them” (Williams 2017). This definition has a broader approach covering indecent or suggestive remarks, unwanted touching, requests or demands for sex and the dissemination of pornography.

Fitzgerald (1990) grouped definitions of sexual harassment into two, “those that are a priori (theoretical) in nature, and [secondly] those that have been developed empirically, particularly through investigation of what various groups of people perceive sexual harassment to be under different circumstances and in different contexts” (see Fitzgerald 1990 for a detailed overview). Definitions of sexual harassment that are behaviorally focused may be more effective for identifying incidences of sexual harassment and developing effective strategies for preventing sexual harassment. Behaviorally oriented definitions of sexual harassment may be most effective when developed by taking into consideration the sociocultural values of the people within a particular organization. Establishing a list of behavioral statements that offer guidance as to what constitutes sexual harassment, while taking into consideration the sociocultural values of the people within the setting, grants organizations, such as universities, plays a critical role in the identification of sexual harassment incidences.

Eastern Mediterranean University (EMU) in North Cyprus is a multicultural institution with close to 19000 students from 106 different countries (www.emu.edu.tr). More than forty percent of the students are coming from Turkey, less than 20 percent of the students are locals, and about forty percent of the students are international. The international students of EMU are from West African, Middle Eastern, and Central Asian countries. Not only the University but also the country is increasingly attracting more international students from across the globe (Gunsen 2014). The rise in the number of students from different cultures increases the need for a cross-culturally accepted understanding of sexual harassment. However, the absence of a clear definition of sexual harassment does exist at EMU.

Therefore, it is the aim of this study to analyze the relationship between gender, age and nationality, and EMU students’ (i) perceptions of sexual harassment behavior and (ii) attitudes toward sexual harassment.

2 Background

Rotundo et al. (2001), in their meta-analytic review of gender differences in perceptions of sexual harassment, found that women perceived a wider range of behaviors as sexual harassment compared to men across seven behavioral categories of sexual harassment. These categories included derogatory attitudes of a personal and impersonal nature, unwanted dating pressure, sexual propositions, physical sexual contact, nonsexual physical contact, and sexual coercion. The authors provide descriptions and behavioral examples for these seven categories of sexual harassment (Rotundo et al. 2001). The definitions are more abstract compared to the behavioral examples. For practical purposes, the behavioral examples may be more useful for identifying an incidence of sexual harassment. Studies in higher education presented different findings. In the United States, Bursik and Gefter (2011) explored undergraduate university students’ perceptions of sexual harassment in 1990 and 2000, with two different student samples. The authors found that students’ perceptions of what constitutes sexual harassment were diverse but appear to have remained stable at the two-time points. More importantly, the authors conclude that differences in personal perceptions of sexual harassment and legal definitions may be damaging to victims and academic institutions (Bursik and Gefter 2011). The authors suggest that further research is needed with more diverse groups, as there were significant differences in perceptions of sexual harassment between Caucasian and non-Caucasian participants in 2000. In her study, Barr (1993) found no differences between genders in perceptions of sexual harassment behavior. In another study of university students, Corr and Jackson (2001) found differences in sexual harassment perceptions of male and female students.

In a cross-cultural study, Sigal and colleagues (2005) investigated perceptions of sexual harassment among university students in the United States, Canada, Germany, Netherlands, Ecuador, Pakistan, the Philippines, Taiwan, and Turkey. Participants were given a scenario containing sexual harassment and were asked to rate the severity of the sexual harassment behavior, to determine whether the perpetrator was guilty and if so, to choose a type of punishment ranging from mild to severe. The authors found that participants in individualistic countries (Unites States, Canada, Germany, and the Netherlands) reported the perpetrator guilty of sexual harassment more often than participants from collectivist countries (Ecuador, Pakistan, the Philippines, Taiwan, and Turkey). These findings demonstrate differences due to cultural factors among people’s perceptions of what constitutes sexual harassment.

As the current literature suggests, there may be differences in people’s perceptions of what constitutes sexual harassment and their attitudes toward sexual harassment. Therefore, this study aims to answer the following research questions:

Research Question 1::

Is there a difference in the perceptions of sexual harassment behavior of university students based on gender, age, and nationality?

Research Question 2::

Does gender, age, or nationality influence attitudes toward sexual harassment?

3 Methods

3.1 Instrument

A questionnaire was designed to collect data. The questionnaire consisted of three parts. In part one, university students’ perceptions of what constitutes sexual harassment behavior were measured by adapting 15 items from Nielson (1996) and Blakely et al. (1995). All fifteen statements were modified for the higher education environment. The second part of the questionnaire was designed to measure university students’ attitudes toward sexual harassment and contained 13 items that were adapted from the Sexual Harassment Attitude Scale (SHAS) (Mazer and Percival 1988; Ford and Donis 1996). The SHAS measures sexist attitudes toward women. The original SHAS that was developed by Mazer and Percival (1988), which contained 25-items, was reduced to 12-items for this study. This is because the 25 questions of the SHAS scale contained general questions about sexual harassment, and so only questions that were relevant to identifying the perceptions of a student sample on sexual harassment were retained. The last question on the original SHAS scale, “women are more emotional and think less clearly than men,” was divided into two items: “women think less clearly than men,” and “women are more emotional than men.” The third part of the questionnaire consisted of demographic questions related to participants’ age, gender, education, and nationality. The 28 items were selected to represent the spectrum of major types of sexual harassment and were randomly ordered in the survey instrument and standardized in terms of length and format to limit lengthier worded items. Previous research shows that studies investigating sexual harassment have tended to exclude behaviors which are perceived to have a low potential to harass and have also provided respondents with limited responses. In attempting to deal with these issues, the survey instrument utilized in this study was designed to measure students’ perceptions of a wide range of actual workplace behaviors which create varying levels of a hostile environment.

For each of the 28 items in part one and two of the survey, participants were instructed to indicate their degree of agreement with statements on a 5-point Likert scale (1 – strongly disagree; 2 – disagree; 3 – neutral; 4 – agree; 5 – strongly agree). The survey items were adapted and prepared in English and then translated into Turkish by using the back translation method. A pilot study was conducted to determine whether the English and Turkish versions of the questionnaire were understood by the participants. After confirming the questions were clear to participants, the researchers distributed the questionnaires to students on the EMU campus. The questionnaires were completed voluntarily and returned to the researcher without gathering any identifying information (i.e., name, student number).

Participants were purposively recruited and given the questionnaires on the EMU campus to yield a diverse sample in terms of gender, education level, and nationality. Potential participants were approached by the researchers on the EMU campus and informed about the purpose of the study.

3.2 Participants

Initially, 510 questionnaires were distributed; however, not all were returned or completed. The final sample consisted of 460 (male n = 293 (64%); female n = 167 (36%)) full-time university students studying in various disciplines at Eastern Mediterranean University (EMU). Students’ ages ranged from 16 to over 30, with the majority of the sample (n = 319) reporting their age between 21–25 (69%) years of age. Four hundred and twenty-five students (92%) were studying for an undergraduate degree, while 8% (n = 35) of the sample were graduate students. Forty-eight percent of the respondents were students from Turkey, 23% were from Nigeria, 19% from North Cyprus (locals) and 10% were from other international countries such as Iran, Iraq, Syria, Russia, Ukraine, Tajikistan, Turkmenistan, Azerbaijan, Cameroon, and Zimbabwe.

4 Findings

4.1 Analysis of Participants’ Perceptions of Sexual Harassment Behavior

To determine the extent of differences among male and female participants’ perceptions of what constitutes sexual harassment behavior, an independent-samples t-test was conducted. Interestingly, of the 15 items measured, only one item, “students looking at other students in a sexual way,” revealed a significant gender difference (p < .05) among participants’ perceptions. Female participants (m = 3.95, SD = 1.016) were more likely than male participants (m = 3.80, SD = 1.118) to agree with this item as a form of sexual harassment, and male participants were more likely to disagree that the aforementioned behavior was a form of sexual harassment. The remaining 14 items showed no gender differences in participants’ (dis)agreement as to whether a behavior was sexual harassment. Apart from one item, male and female participants appeared to share similar perceptions of behaviors that could be classified as sexual harassment.

To determine if respondents’ perceptions of sexually harassing behavior change as students grow older, two age groups -younger (16–23 years old) and older (24+ years old) students- were formed. It was found that younger and older student groups perceived two items “students calling others names like ‘slut,’ ‘whore,’ etc.” and “students forcing other students to have sex” differently (p < .01, p < .05 respectively). The older student group (m = 3.63, SD = 1.218) were more likely than the younger student group (m = 3.52, SD = 1.153) to agree that “students calling others names like ‘slut,’ ‘whore,’ etc.” was sexually harassing behavior (p < 0.01). The older student group (m = 3.92, SD = 1.144) were more likely than the younger student group (m = 3.86, SD = 1.115) to agree that “students forcing other students to have sex” was sexually harassing behavior (p < .05). These findings suggest that some behaviors perceived as sexual harassment by older students may not be perceived as sexual harassment by younger students. In other words, younger students may have a narrower perception of which behaviors may be sexual harassment compared to older participants.

The effect of culture on the perceptions of EMU students on sexually harassing behavior was also examined. Respondents were grouped based on their nationalities as students from Turkey, North Cyprus, and International students. ANOVA results indicated that students from different cultures perceived three non-verbal behavior items (“Students showing, giving or leaving sexually offensive pictures, photos or messages for other students,” “Students grabbing or sexually pinching other students,” and “Students touching or pinching other students’ private parts”) significantly different (at p < .05 for three items). Students from Turkey and North Cyprus were more likely than international students to agree on non-verbal behaviors to be sexually harassing behavior. International students in this sample may be less likely to perceive behaviors, particularly non-verbal ones, as sexual harassment compared to students from Turkey and North Cyprus.

4.2 Analysis of Participants’ Attitudes Toward Sexual Harassment

The second part of the questionnaire included statements of attitude toward sexual harassment. For the second part of the questionnaire, responses were compiled to obtain a single score of indicating participants’ tolerance for sexual harassment. This was achieved by summing the scores of statements endorsing the view that sexual harassment is not a problem, and women should expect such behavior in their environment. Lower scores indicate low tolerance of sexual harassment, and higher scores indicate higher tolerance for sexual harassment. Here, data from 36 male respondents and 11 female respondents were eliminated from further analysis due to incompleteness, so the total sample for this analysis was 413 (female = 156; male = 257). Results of the mean scores of tolerance for sexual harassment by gender show that female students have a lower tolerance (m = 37.90) for sexual harassment than male students (m = 42.29), indicating that female students were more likely to view the statements listed in the questionnaire as sexual harassment. These findings indicate that female participants more often reported agreeing that the statements listed were potentially sexual harassment compared to the agreement from male participants.

Further analyses were conducted to measure whether sexual harassment attitudes differ between male and female respondents. According to t-test results, male and female EMU students perceived 7 out of 13 items of attitudes of sexual harassment significantly different (at p < .05). Female EMU students agreed more than male EMU students that “It is only natural for a woman to use her sexuality as a way of getting ahead” (at p < 0.00), “Innocent flirtations make the day interesting” (at p < 0.01), “Encouraging an instructor’s/assistant’s sexual interest is frequently used by women to improve their grades” (at p < 0.01), “A lot of what people call sexual harassment is just normal flirtation” (at p < 0.01), “All this concern about sexual harassment makes it harder for men and women to have normal relations” (at p < .05), “Women think less clearly than men” (at p < .05), “Women are more emotional than men” (at p < .00) are sexually harassing attitudes.

ANOVA results also showed that there were statistically significant differences between younger EMU students (16–23 years old) and older students (24+ years old) for six items at p < .05. Younger university students more often reported agreeing with sexually harassing attitudes, such as “It is only natural for a woman to use her sexuality as a way of getting ahead”, “A lot of what people call sexual harassment is just normal flirtation”, “Encouraging an instructor’s/assistant’s sexual interest is frequently used by women to improve their grades”, “One of the problems with sexual harassment is that women cannot take a joke”, “All this concern about sexual harassment makes it harder for men and women to have normal relations”, “Women think less clearly than men”, and “Women are more emotional than men” than older EMU students. The findings suggest that age may be an important factor in determining participants’ attitudes toward sexual harassment. Older participants were less likely to agree with the statements endorsing the view that sexual harassment is not a problem compared to younger participants. Since there were no statistically significant differences in the attitudes of students toward sexual harassment based on nationality, we need to indicate that our findings do not support that culture is an important factor in determining participants’ attitudes toward sexual harassment.

5 Discussion

This study aimed to identify university students’ perceptions of behaviors that constitute sexual harassment and compare university students’ sexual harassment attitudes based on gender, age, and nationality. The findings suggest that while there appears to be some agreement as to which behaviors constitute sexual harassment based on gender, there may be some differences in attitudes toward sexual harassment based on participants’ gender. EMU students’ perceptions were similar toward what constitutes sexually harassing behavior, regardless of gender. However, women were more likely than men to perceive a wider range of verbal and non-verbal behaviors as sexual harassment. This is consistent with previous findings by Rotunda and colleagues (2001). In support with previous research, results from this study indicate that female students perceive more behaviors to be indicative of sexual harassment compared to male students (Gutek et al. 1980; Popovich et al. 1986). Results also show a significant difference between male and female students’ attitudes toward sexual harassment. Females who participated in the study have significantly less tolerance for sexual harassment than men. Such results may be related to the fact that men, in general, tend to be the perpetrators, whereas women tend to be the victims of sexual harassment (Ford and Donis 1996). Another possible explanation for females’ lower tolerance is that they may be viewed as targets by harassers, thus making them more sensitive to sexual harassment (Blakely et al. 1995). These findings are similar to those of Ford and Donis (1996) who showed gender to be an important factor in determining attitudes toward sexual harassment, revealing a significant effect of age on perceived differences of sexual harassment attitudes. This is consistent with the findings of this study, as older respondents’ tolerance was found to be lower than younger respondents, indicating that older respondents were more likely to view an item as indicative of sexual harassment compared to younger respondents.

The perceptions of sexual harassment behavior and attitudes toward sexual harassment based on the age of participants indicated differences. It is found that younger participants have a narrower perception of which behaviors constitute sexual harassment compared to older participants. Interestingly, younger participants were more likely to agree with sexists attitudes compared to older participants. As this was a cross-sectional design, it is not possible to determine whether this is a cohort effect or whether attitudes change over time. A longitudinal design would be needed to examine this premise.

The perceptions of sexual harassment behavior indicated differences by participants’ nationality. A novel contribution of this study was that international participants, compared to participants from North Cyprus and Turkey, appeared to be more tolerant of which behaviors relate to sexual harassment. In other words, students from North Cyprus and Turkey appeared to be more likely to classify more behaviors as sexual harassment compared to students from other cultures.

The findings presented in this study may seem debatable, given the differences in the results for perceptions of potentially sexual harassment behaviors, and potentially sexual harassment attitudes. However, this is not unusual, as research examining gender differences in perceptions of what constitutes sexual harassment has presented mixed results. Researchers have argued that this may be due to the degree of ambiguity in the examples of the sexually oriented items being measured (Blakely et al. 1995). Offensive behaviors may be unambiguous that disagreement about its interpretation is highly unlikely. This could be the reason why the results for respondents’ perceptions of potentially harassing sexual behaviors in this study showed no significant difference, while that of potentially harassing sexual attitudes showed a significant difference. Here, significant differences were found because they asked less specific questions that provided limited contextual information.

The findings from the current study are consistent with the literature, which suggests that in academic settings, both male and female perceptions of sexual harassment varied for ambiguous scenarios. When the items were highly explicit, there were higher levels of agreement between males and females; however, when the items were ambiguous, females were more likely than males to perceive the behavior as sexually harassing (Adams et al. 1983; Kenig and Ryan 1986).

It is clear that organizations have a responsibility to provide an educational environment free of sexual harassment. Pro-active steps should take into consideration gender-, age- and culture-based differences to fulfill EMU’s responsibility in this regard. Specific areas that EMU might take action include training of employees and student assistants (Thacker and Gohmann 1993). Equally important is the need for EMU to investigate harassment claims quickly and thoroughly, taking appropriate disciplinary steps. Organizations that have failed to act upon claims of sexual harassment risk finding that their negligence contributed to increasing the organization’s liability. Being able to understand the nature of gender-, age- and culture-based differences can do much to reduce confusion in this area.

It is recommended that the first step in eliminating sexual harassment is to develop an organizational policy that clearly states that sexual harassment will not be tolerated, providing examples of behaviors (verbal and non-verbal) that will be interpreted as sexual harassment, and the consequences of not adhering to the policy. EMU’s hierarchical organization would benefit from involving students in the policy-making process since sexual harassment is often observed in relationships of unequal power (MacKinnon 1979). It would be desirable to include male and female student perceptions of different ages and cultures in the policy-making process.

Another recommendation is to educate employees and students about the policy, to increase sensitivity to behaviors that may be perceived to be sexual harassment. As part of sexual harassment prevention training, organizations should sensitize employees at all levels of the organization to the various forms of hostile environment harassment. Stressing the fact that it is the victim’s perception and not the intent of the accused that will be taken into consideration, may encourage members of the organization to think about how the recipient of their actions may perceive their behaviors.

6 Conclusion

This study presents information gathered from a diverse sample of students at EMU about their views on behaviors and perceptions about sexual harassment. These findings suggest that while there appears to be some agreement as to which behaviors constitute sexual harassment, there may be some differences in people’s perceptions of sexual harassment related to their gender, culture, and age. The findings indicate that female participants more often reported agreeing that the behaviors listed were potentially sexual harassment compared to male participants. Age is found to be an important factor both in participants’ perceptions of sexual harassment behavior and in their attitudes toward sexual harassment. Younger participants may have a narrower perception of which behaviors constitute sexual harassment compared to older participants. Similarly, when analyzing attitudes, older participants were less likely to agree with statements endorsing sexist attitudes. Another important finding of the study was that international participants, compared to participants from North Cyprus and Turkey, appeared to be more tolerant of which behaviors constitute sexual harassment.

The findings from this study lay the foundations for developing guidelines and policies on sexual harassment at EMU. While it is desirable for organizations to establish specific definitions of sexual harassment, educating people within an organization to identify or prevent sexual harassment is essential for effectively dealing with sexual harassment cases. Without a commonly accepted, behaviorally-based definition of what constitutes sexual harassment, the degree to which the problem exists within an organization cannot be accurately assessed, and the problem cannot be effectively addressed. Organizations, such as academic institutions, need to establish clear sexual harassment policies and procedures to effectively prevent, identify, and handle sexual harassment cases (Witkowska and Menckel 2005). Without a clear behavioral–based definition of what constitutes sexual harassment, sexual harassment cases may not be effectively prevented, evaluated, and managed. From a practical perspective, the findings could be used to establish a list of behaviors that could serve as a guideline for establishing sexual harassment cases.

A limitation of this study was that participants’ personal experiences of sexual harassment on the school premises, the types of harassment they may have experienced, and their views of whether sexual harassment is a problem at the institution were not collected. Such information would have been useful for the researchers in making comparisons between age, gender, and culture to examine potential relationships with the behaviors identified as sexual harassment and students’ experiences. However, the researchers did not want to risk causing distress for any participants as a result of completing the survey. Another limitation may be that participants were responding in socially desirable ways, which may explain why no significant difference was found in male and female perceptions of sexual harassment behaviors. This study’s use of a Likert style response scale may aide the identification of behaviors and perceptions that could be creating and sustaining a hostile school environment. For example, students may agree with the statement “Most of the men are sexually teased by women whom they interact,” without considering this as sexual harassment. Though such behaviors are not typically reported, such behaviors may hurt the victim’s productivity, absenteeism, and career success.

Future research could investigate perceptions of sexual harassment using a more qualitative approach. This study was limited in terms of the items participants were asked to respond to, and therefore, it is likely that many other relevant behaviors and attitudes about sexual harassment may have been missed. Interviews with a diverse sample of participants in terms of age, gender, and nationality could facilitate establishing a more encompassing set of guidelines for preventing, identifying, and managing sexual harassment incidences.